LLM Stats (PostTrainBench): leaderboard
PostTrainBench as reported by LLM Stats — comparison of models on post-training and adaptation tasks; leaderboard score.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around 2029. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 45.3 | #8 |
| 2 | GLM-5.3 | 39.8 | #39 |
| 3 | MiniMax-M3.1-Flash | 39.3 | #33 |
| 4 | MiniMax-M3 | 37.1 | #136 |
| 5 | Kimi K3 | 36.6 | #30 |
| 6 | Hy4 preview | 35.6 | #38 |
| 7 | GLM-5.2 | 34.3 | #73 |
| 8 | Seed 2.1 Turbo | 18.3 | #86 |
| 9 | Seed 2.1 Pro | 16.5 | #67 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-posttrainbench · How It Works · Data refreshed daily, snapshot 2026-10-20.